Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervision
學習人臉反欺騙的深度模型：二進制或輔助監控 摘要 人臉反欺騙是防止人臉識別系統的安全漏洞的關鍵。以往的深度學習方法將人臉反欺騙表述爲一個二值分類問題。他們中的許多人很難掌握足夠的欺騙線索，因而無法進行有效的歸納。在本文中，我們認爲輔助監督對於引導朝着區分性和普遍性限線索的學習的很重要。利用CNN-RNN模型對人臉深度進行像素監督估計對rPPG信號進行序列監督評估。估計深度和rPPG信號被融合來區分
相關文章
- 1. 單目可見光靜默活體檢測 Binary or Auxiliary Supervision論文解讀
- 2. 《Learning Deep Representation for Face Alignment with Auxiliary Attributes》論文閱讀筆記
- 3. A Discriminative Feature Learning Approach for Deep Face Recognition
- 4. Deep Learning Models and Code for Pose Estimation
- 5. Deep Learning Models on Kubernetes with GPUs
- 6. 論文：Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots
- 7. Deploying deep learning models in OpenCV
- 8. Learning Invariant Deep Representation for NIR-VIS Face Recognition
- 9. awesome-face-anti-spoofing
- 10. Deep Learning for NLP Best Practices
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • Java Agent入門實戰（三）-JVM Attach原理與使用
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 單目可見光靜默活體檢測 Binary or Auxiliary Supervision論文解讀
- 2. 《Learning Deep Representation for Face Alignment with Auxiliary Attributes》論文閱讀筆記
- 3. A Discriminative Feature Learning Approach for Deep Face Recognition
- 4. Deep Learning Models and Code for Pose Estimation
- 5. Deep Learning Models on Kubernetes with GPUs
- 6. 論文：Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots
- 7. Deploying deep learning models in OpenCV
- 8. Learning Invariant Deep Representation for NIR-VIS Face Recognition
- 9. awesome-face-anti-spoofing
- 10. Deep Learning for NLP Best Practices